Related Experiment Video
Updated: Jun 28, 2025
09:16
Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis
Published on: June 18, 2020
6.9K
Using Deep Learning to Automate Eosinophil Counting in Pediatric Ulcerative Colitis Histopathological Images
Medrxiv : the Preprint Server for Health Sciences
|April 18, 2024
Summary
This study developed an automated deep learning tool for counting eosinophils in ulcerative colitis biopsies, achieving high accuracy and concordance with expert pathologists. This AI tool offers a reliable alternative to manual cell counting in histopathology.
Area of Science:
- Histopathology
- Artificial Intelligence
- Gastroenterology
Background:
- Accurate identification of inflammatory cells, particularly eosinophils, in mucosal histopathology is crucial for diagnosing ulcerative colitis (UC).
- Eosinophil counts in colonic mucosa correlate with UC disease course.
- Manual cell counting is time-consuming and prone to subjective bias.
Conclusions:
- Deep learning-based automated eosinophilic cell counting provides a robust and accurate method for analyzing histopathology images.
- The developed tool shows a high degree of concordance with manual expert annotations, offering a reliable alternative for clinical practice.
